Phylogenetic Tree

Load packages

library(tidyverse)
library(gridExtra)
library(cowplot)
library(ggridges)
library(ggstance)
library(treeio)
library(ggtree)
library(tidytree)

Load and combine data

species <- read_csv("../data/species_data.csv") %>% select(-species)
data <- read_csv("../data/data.csv") %>% 
  left_join(species, by = c("species" = "species_formatted")) %>% 
  rename(label = species_latin)
data2 <- data %>% group_by(species, label) %>% summarise(n = sum(n))
fulltree <- read.nexus("../data/consensusTree_10kTrees_298Primates_V3.nex")
refs <- read_csv("../data/ref_nodes.csv")

# turn tree into tidy dataframe
tree2 <- as_tibble(fulltree)

tree3 <- tree2 %>% 
  left_join(data2) %>% 
  mutate(
    hasN = ifelse(is.na(n), 0, .5),
    hasN2 = ifelse(is.na(n), .1, .5)) %>% 
  left_join(refs) %>% 
  # # also merge w datasheet listing node, n for inner nodes
  # left_join(Ns, by = "node") %>% 
  # mutate(n = coalesce(n.x, n.y)) %>% 
  # select(-n.x, -n.y) %>% 
  groupClade(c(493, 496, 429, 302, 408)) %>% 
  mutate(group = fct_recode(group, "2" = "1"))

Joining, by = "label"
Joining, by = "node"

# turn back into tree
tree4 <- as.treedata(tree3)

Figure out nodes

This makes a rectangular and a circular tree with the node numbers displayed for reference (saved in the graphs folder).

tree3.2 <- as.treedata(tree3)

# display node numbers for reference
ggtree(tree3.2) +
  # tip labels
  geom_tippoint(aes(size = n), col = "seagreen", alpha = .5) +
  geom_tiplab(offset = 1, size = 3) +
  # node labels
  geom_text(aes(label = node, x = branch), size = 2, col = "blue", vjust = -.5) +
  expand_limits(x = 90) +
  # display timescale at the bottom
  theme_tree2()

ggsave("../graphs/full_tree_nodes.pdf", width = 8, height = 20, scale = 2)

ggtree(tree3.2, layout = "circular") +
  geom_tippoint(aes(size = n), col = "seagreen", alpha = .5) +
  geom_tiplab2(offset = 2, size = 3) +
  geom_text2(aes(label = node), size = 1.5, col = "blue") +
  xlim(NA, 100)

ggsave("../graphs/full_tree_nodes_circular.pdf", width = 8, height = 8, scale = 2)

Circular tree of 298 primates

cols <- viridis::viridis(4, end = .9)

# base plot
p <- ggtree(tree4, aes(size = hasN, alpha = hasN2), layout = "circular") +
  # root
  geom_rootpoint(size = 1) +
  # tips
  geom_tippoint(aes(size = n), alpha = .5) +
  geom_tiplab2(aes(alpha = hasN), offset = 2, size = 3) +
  # tweak scales
  scale_alpha_continuous(range = c(.3, 1)) +
  scale_size_continuous(range = c(.5, 8)) +
  # widen plotting area
  xlim(NA, 100)

# highlight clades with background colors
p2 <- p + 
  geom_hilight(node = 493, fill = cols[1], alpha = .3) +
  geom_hilight(node = 496, fill = cols[1], alpha = .3) +
  geom_hilight(node = 429, fill = cols[2], alpha = .3) +
  geom_hilight(node = 303, fill = cols[3], alpha = .3) +
  geom_hilight(node = 408, fill = cols[4], alpha = .3) +
  # plot tree again to be on top of the highlights
  geom_tree() +
  geom_rootpoint(size = 1)

# highlight clades with branch colors
p3 <- p + 
  aes(col = group) +
  scale_color_manual(values = c("gray30", cols))

plots <- mget(c("p", "p2", "p3"))

grid.arrange(p, p2, p3, nrow = 1)

# png with 3x1
ggsave("../graphs/phylo_full.png", arrangeGrob(grobs = plots, nrow = 1), 
       width = 24, height = 8, scale = 2, dpi = 72)

Removed 287 rows containing missing values (geom_point_g_gtree).Removed 287 rows containing missing values (geom_point_g_gtree).Removed 287 rows containing missing values (geom_point_g_gtree).

# pdf with 1 per page
ggsave("../graphs/phylo_full.pdf", marrangeGrob(grobs = plots, nrow = 1, ncol = 1), 
       width = 8, height = 8, scale = 2, dpi = 72)

Removed 287 rows containing missing values (geom_point_g_gtree).Removed 287 rows containing missing values (geom_point_g_gtree).Removed 287 rows containing missing values (geom_point_g_gtree).

Sample size in detail

# subset tree to just those species who have sample sizes reported, i.e. those who were tested
to_drop <- tree3 %>% filter(is.na(n)) %>% pull(label)
tree5 <- drop.tip(tree4, to_drop)

data3 <- data %>% 
  select(label, everything()) %>% 
  rename(num = n)

# species with more than X sites can get a density?
d3a <- data3 %>% group_by(species) %>% filter(n_distinct(site) >= 4)
d3b <- data3 %>% # setdiff(data3, d3a) ## <- do setdiff instead to NOT show points for densities
  group_by(species) %>% 
  # create variable num2 is NA if there's only one data point for a species
  # --> those species will only get the vertical crossbar
  mutate(flag = n_distinct(site) == 1) %>% 
  ungroup %>% 
  mutate(num2 = ifelse(flag, NA, num))

# for vertical crossbar = median
data4 <- data3 %>% 
  group_by(label, species) %>% 
  summarise(Mdn = median(num)) # totalN = sum(num), sitesN = n_distinct(site)

q <- ggtree(tree5, aes(col = group)) +
  # this is a dummy point to expand the x scale
  # the typical ways weirdly also expand it for the sample size panel once that's added
  geom_point(data = tibble(x = 135, y = 1, .panel = "Tree", group = NA)) +
  # tip labels
  geom_tippoint(aes(size = n), alpha = .5) +
  geom_tiplab(offset = 4, size = 3) +
  # tweak scales
  scale_color_manual(values = c("grey30", cols)) +
  scale_fill_manual(values = cols[4]) + # when all categories are taken: cols
  # display timescale at the bottom
  theme_tree2()

# trying out which geoms to layer
s1 <- ggplot(data3, aes(x = num, y = label, group = label)) +
  geom_density_ridges(lwd = .3, col = "grey80", bandwidth = 1)

s2 <- ggplot(data3, aes(x = num, y = label, group = label)) +
  geom_crossbarh(data = data4, aes(x = Mdn, xmin = Mdn, xmax = Mdn), width = .5) +
  geom_point(alpha = .5)

s3 <- ggplot(data3, aes(x = num, y = label, group = label)) +
  geom_boxploth()

plot_grid(s1, s2, s3, nrow = 1)

# right-side viz depends on the number of sites per species:
# 1 site = vertical crossbar only
# 2+ sites = points + crossbar at median
# X+ sites = densities (currently, X = 4 just to illustrate)

# dirty hack: there's a small character in front of "Sample sizes" to have that panel sort to the right (alphabetically) until I figure out why it doesn't just go by order. This cropped up as an issue when I added the dummy point for the x-axis expansion...

q %>% 
  facet_plot("ٴ Sample sizes", 
             d3a, geom_density_ridges, aes(x = num, group = label, fill = group), 
             alpha = .5, lwd = .3) %>% 
  facet_plot("ٴ Sample sizes",
             data4, geom_crossbarh, aes(x = Mdn, xmin = Mdn, xmax = Mdn, group = label, 
             col = group), alpha = .5, width = .3, position = position_nudge(y = -.1)) %>%
  facet_plot("ٴ Sample sizes", 
             d3b, geom_point, aes(x = num2, group = label), alpha = .5, 
             position = position_nudge(y = -.1)) +
  panel_border()

ggsave("../graphs/phylo_ridge.png", width = 4, height = 3, scale = 2)

Relevant functions to tinker with, if necessary:

facet_plot

function (p, panel, data, geom, mapping = NULL, ...) 
{
    p <- add_panel(p, panel)
    df <- p %+>% data
    p + geom(data = df, mapping = mapping, ...)
}
<bytecode: 0x7f8a0db5f978>
<environment: namespace:ggtree>

ggtree:::add_panel

function (p, panel) 
{
    df <- p$data
    if (is.null(df[[".panel"]])) {
        df[[".panel"]] <- factor("Tree")
    }
    levels(df$.panel) %<>% c(., panel)
    p$data <- df
    p + facet_grid(. ~ .panel, scales = "free_x")
}
<bytecode: 0x7f8a0db62660>
<environment: namespace:ggtree>

Session info

sessionInfo()

R version 3.6.1 (2019-07-05)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS Mojave 10.14.5

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] tidytree_0.2.8  ggtree_1.16.6   treeio_1.8.2    ggstance_0.3.3  ggridges_0.5.1 
 [6] cowplot_1.0.0   gridExtra_2.3   forcats_0.4.0   stringr_1.4.0   dplyr_0.8.3    
[11] purrr_0.3.2     readr_1.3.1     tidyr_1.0.0     tibble_2.1.3    ggplot2_3.2.1  
[16] tidyverse_1.2.1

loaded via a namespace (and not attached):
 [1] tidyselect_0.2.5   xfun_0.10          reshape2_1.4.3     haven_2.1.1        lattice_0.20-38   
 [6] colorspace_1.4-1   vctrs_0.2.0        generics_0.0.2     viridisLite_0.3.0  rlang_0.4.0       
[11] pillar_1.4.2       glue_1.3.1         withr_2.1.2        modelr_0.1.5       readxl_1.3.1      
[16] rvcheck_0.1.5      lifecycle_0.1.0    plyr_1.8.4         munsell_0.5.0      gtable_0.3.0      
[21] cellranger_1.1.0   rvest_0.3.4        labeling_0.3       knitr_1.25         parallel_3.6.1    
[26] broom_0.5.2        Rcpp_1.0.2         BiocManager_1.30.7 scales_1.0.0       backports_1.1.5   
[31] jsonlite_1.6       hms_0.5.1          stringi_1.4.3      grid_3.6.1         cli_1.1.0         
[36] tools_3.6.1        magrittr_1.5       lazyeval_0.2.2     crayon_1.3.4       ape_5.3           
[41] pkgconfig_2.0.3    zeallot_0.1.0      xml2_1.2.2         lubridate_1.7.4    viridis_0.5.1     
[46] assertthat_0.2.1   httr_1.4.1         rstudioapi_0.10    R6_2.4.0           nlme_3.1-141      
[51] compiler_3.6.1